TWI843531B - Memory device and method for forming the same - Google Patents

Abstract

A memory device includes a first pull-down transistor, a first pass-gate transistor, a second pull-down transistor, a second pass-gate transistor, a first pull-up transistor, and a second pull-up transistor. A first power line, a first bit line, and a second bit line is provided, the first power line includes first and second portions separated from each other, wherein in a cross-sectional view, the second portion of the first power line is laterally between the first and second bit lines along a direction. A first via electrically connects the first portion of the first power line to the first pull-down transistor. A second via electrically connects the first bit line to the first pass-gate transistor. A third via electrically connects the second portion of the first power line to the second pull-down transistor. A fourth via electrically connects the second bit line to the second pass-gate transistor.

Description

Memory element and method of forming the same

This disclosure relates to a memory device and a method for forming the same.

The semiconductor integrated circuit (IC) industry has experienced rapid growth, with each generation of circuits being smaller and more complex than the previous generation. However, these advances have also been accompanied by increases in process complexity, and similar IC process development is necessary to achieve these advances.

In the evolution of integrated circuits, functional density (such as the number of interconnected components per chip) has been increasing, while component size (such as the smallest component that can be manufactured by the process) has been shrinking. The shrinking process generally provides improved production efficiency and reduces the associated waste. The reduction in size also increases the relatively high power consumption, which can be solved by using low-power components (such as complementary metal oxide semiconductor components).

According to some embodiments of the present disclosure, a memory element includes a substrate. A first pull-down transistor, a first gate transistor, a second pull-down transistor, and a second gate transistor are located at a first level above the substrate. A first pull-up transistor and a second pull-up transistor are located at a second level above the first level. A first power line, a first bit line, and a second bit line, the first power line including a first portion and a second portion separated from each other, wherein in a cross-sectional view, the second portion of the first power line is laterally located between the first bit line and the second bit line along a direction. A first conductive via contacts an upper surface of the first portion of the first power line and is electrically connected to a source/drain region of the first pull-down transistor. A second conductive via contacts an upper surface of the first bit line and is electrically connected to a source/drain region of the first gate transistor. A third via contacts the upper surface of the second portion of the first power line and is electrically connected to the source/drain region of the second pull-down transistor. A fourth via contacts the upper surface of the second bit line and is electrically connected to the source/drain region of the second gate transistor, wherein in a top view, the first via is laterally aligned with the fourth via along the direction, and the second via is laterally aligned with the third via along the direction.

According to some embodiments of the present disclosure, a memory device includes a substrate. A first pull-down transistor, a first gate transistor, a second pull-down transistor, and a second gate transistor are located at a first level above the substrate. A first pull-up transistor and a second pull-up transistor are located at a second level above the first level, wherein a gate structure of the first pull-up transistor vertically overlaps a gate structure of the first pull-down transistor, and a gate structure of the second pull-up transistor vertically overlaps a gate structure of the second pull-down transistor. A first power line, a first bit line, and a second bit line, wherein the first power line includes a first portion and a second portion separated from each other. A first conductive via contacts an upper surface of a first portion of the first power line and is electrically connected to a source/drain region of the first pull-down transistor. The second conductive via contacts the upper surface of the second portion of the first power line and is electrically connected to the source/drain region of the second pull-down transistor. The third conductive via contacts the upper surface of the first bit line and is electrically connected to the source/drain region of the first gate transistor. The fourth conductive via contacts the upper surface of the second bit line and is electrically connected to the source/drain region of the second gate transistor. The first liner, the second liner, the third liner, and the fourth liner are respectively attached to the sidewall of the first portion of the first power line, the sidewall of the second portion of the first power line, the sidewall of the first bit line, and the sidewall of the second bit line. The first dielectric material, the second dielectric material, the third dielectric material and the fourth dielectric material are respectively lined with the sidewalls of the first via hole, the sidewalls of the second via hole, the sidewalls of the third via hole and the sidewalls of the fourth via hole, wherein the first liner, the second liner, the third liner and the fourth liner are composed of materials different from the first dielectric material, the second dielectric material, the third dielectric material and the fourth dielectric material.

According to some embodiments of the present disclosure, a method for forming a memory device includes forming a first semiconductor strip and a second semiconductor strip protruding above a substrate. Forming a first isolation structure, a second isolation structure, and a third isolation structure above the substrate, wherein the first semiconductor strip is located between the first isolation structure and the second isolation structure, and the second semiconductor strip is located between the second isolation structure and the third isolation structure. Patterning the second isolation structure and the third isolation structure to form a first trench and a second trench in the second isolation structure and the third isolation structure, respectively. Forming a first metal line and a second metal line in the first trench and the second trench, respectively. After forming the first metal line and the second metal line, patterning the first isolation structure and the second isolation structure to form a third trench and a fourth trench in the first isolation structure and the second isolation structure, respectively. Forming a third metal line and a fourth metal line in the third trench and the fourth trench, respectively. Forming a first pull-down transistor electrically connected to the third metal line, a first gate transistor electrically connected to the first metal line, a second pull-down transistor electrically connected to the fourth metal line, and a second gate transistor electrically connected to the second metal line. Forming a first pull-up transistor and a second pull-up transistor vertically stacked above the first pull-down transistor, the first gate transistor, the second pull-down transistor, and the second gate transistor.

10:SRAM components

90: Substrate

102,104: Opposite device

103,105: Storage node/output port

110,111:Semiconductor layer

112,114,116,118: Semiconductor channel layer

125: Gate spacer

126:Internal partition

130: Dummy gate dielectric

131: Virtual gate electrode

132,134,136,138: Virtual gate structure

150,185: Interlayer dielectric layer

152,154,156,158: Gate structure

171,172,173,174,175,176: Source/drain contacts

180,181:Semiconductor layer

191,192,193,194,195,196: Intermetallic dielectric layer

210,211:Semiconductor layer

212,214:Semiconductor channel layer

230: Dummy gate dielectric

231: Virtual gate electrode

225: Gate spacer

226:Internal partition

232,234: Virtual gate structure

252,254: Gate structure

271,272,273,274: Source/drain contacts

300: Gate dielectric layer

302: Work function metal layer

304:Filled metal

312,314: Semiconductor strips

315,316,317: Isolation structure

321,322,323,324: Lining

331,332,333,334: Dielectric materials

A-A,B-B: line

BL,BLB:bit line

FN11, FN12, FN21, FN22: Fin structure

D1,D2,D3:Distance

LV1: First level

LV2: Second level

MA1, MA2: Mask

ML11,ML12,ML13,ML14,ML31,ML32,ML41,ML51,ML52:Metal wire

MV11,MV12,MV13,MV14,MV21,MV22,MV23,MV24,MV31,MV32,MV33,MV34,MV41,MV42,MV51,MV52,MV201,MV202,MV301,MV302,MV401,MV402,MV403,MV404,MV501,MV502,MV503,MV504,MV601,MV602,MV603,MV604,MV703,MV704,MV803,MV804: Metal vias

O1, O2, O3: Opening

PD-1, PD-2: Pull-down transistor/transistor

PG-1, PG-2: Gate transistor/transistor

PU-1,PU-2:Pull-up transistor/transistor

SD11,SD12,SD13,SD14,SD15,SD16,SD21,SD22,SD23,SD24: Source/Drain epitaxial structure

TR1,TR2,TR3: Grooves

Vdd, Vss: power line

WL: character line

The aspects of the present disclosure are best understood from the following detailed description when read in conjunction with the accompanying drawings. Note that, in accordance with standard practice in the industry, the various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or reduced for clarity of discussion.

Figure 1 is a circuit diagram of a six-transistor SRAM device of some embodiments of the present disclosure.

Figures 2A, 2B and 2C are three-dimensional diagrams of SRAM devices of some embodiments of the present disclosure.

Figure 2D and Figure 2E are cross-sectional views of SRAM devices of some embodiments of the present disclosure.

Figures 3 to 31 show the various stages of a method for forming an SRAM element in some embodiments of the present disclosure.

Figure 32 is a three-dimensional diagram of an SRAM device of some embodiments of the present disclosure.

The following disclosure provides many different embodiments or examples for implementing different features of the subject matter provided. Specific examples of components and configurations are described below to simplify the disclosure. Of course, these are merely examples and are not intended to be limiting. For example, in the following description, the formation of a first feature above or on a second feature may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which additional features may be formed between the first and second features so that the first and second features may not be in direct contact. In addition, in various examples, the disclosure may repeatedly refer to numbers and/or letters. This repetition is for the purpose of simplicity and clarity and does not in itself dictate the relationship between the different embodiments and/or configurations discussed.

Additionally, for ease of description, spatially relative terms such as "beneath," "below," "lower," "above," and "upper," and the like, may be used herein to describe the relationship of one element or feature to another element or feature as illustrated in the figures. These spatially relative terms are intended to encompass different orientations of the elements in use or operation in addition to the orientation depicted in the figures. The device may be otherwise oriented (rotated 90 degrees or in other orientations), and likewise the spatially relative descriptors used herein may be interpreted accordingly.

As used herein, "approximately", "substantially" or "substantially" may mean within 20%, within 10%, or within 5% of a given value or range. However, those skilled in the art will understand that the values or ranges cited throughout this specification are examples only and may decrease as the size of the integrated circuit is reduced. The values given herein are approximate, meaning that if not explicitly stated, they may represent "approximately", "substantially" or "substantially".

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by a person of ordinary skill in the art to which this disclosure belongs. It should also be understood that, unless explicitly defined, those terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and this disclosure, and will not be interpreted in an idealized manner or in an overly formal sense.

A gate all around (GAA) transistor structure can be patterned by any suitable method. For example, the structure can be patterned using one or more photolithography processes, including double patterning or multiple patterning processes. Typically, the double patterning or multiple patterning process combines photolithography and self-alignment processes, which allows for patterns with a pitch smaller than a single direct photolithography process pitch. For example, in one embodiment, a sacrificial layer is formed over a substrate and patterned using a photolithography process. Spacers are formed next to the patterned sacrificial layer using a self-alignment process. The sacrificial layer is removed, and the remaining spacers can then be used to pattern the gate all around structure.

The present disclosure relates to integrated circuit (IC) structures and methods for forming the same. More specifically, some embodiments of the present disclosure relate to gate-all-around components. The gate-all-around component includes a component in which a gate structure (or a portion thereof) is formed on four sides of a channel region (e.g., surrounding a portion of the channel region). The channel region of the gate-all-around component may include a nanosheet channel, a strip channel, and/or other suitable channel configurations. In some embodiments, the channel region of the gate-all-around component may have a plurality of vertically spaced horizontal nanosheets or horizontal strips, such that the gate-all-around component becomes a stacked horizontal gate-all-around (stacked horizontal GAA; S-HGAA) component. The gate full-surround device described herein includes a p-type metal oxide semiconductor gate full-surround device and an n-type metal oxide semiconductor gate full-surround device stacked together. In addition, the gate full-surround device can have one or more channel regions (e.g., nanosheets) associated with a single, continuous gate structure or multiple gate structures. A person of ordinary skill in the art can understand that other exemplary semiconductor devices can also benefit from the different aspects of the present disclosure. In some embodiments, the nanosheets can be alternatively referred to as nanowires, nanoplates, nanorings, or nanostructures with nanoscale dimensions (e.g., a few nanometers), depending on their geometry. In addition, the embodiments disclosed herein can also be applied to various metal oxide semiconductor transistors (e.g., complementary field effect transistors (CFETs) and fin field effect transistors (FinFETs)).

Some embodiments discussed herein are discussed in the context of nanofield effect transistors formed using a gate-last process. In other embodiments, a gate-first process may be used. In addition, aspects of some embodiments may be applied to planar components such as planar field effect transistors or to fin field effect transistors (FinFETs). For example, a fin field effect transistor may include a fin on a substrate, the fin serving as a channel region of the fin field effect transistor. Similarly, a planar field effect transistor may include a substrate, a portion of which serves as a channel region of the planar field effect transistor.

This disclosure will be discussed through the embodiments described in the context, where a static random access memory (SRAM) forms a full gate-around configuration. However, the embodiments of this disclosure can also be applied to various semiconductor devices. Different embodiments will be described below with different figures.

Static random access memory (SRAM) is a volatile semiconductor memory that uses a bi-stable latch circuit to store bits. Bits in SRAM are stored on four transistors (transistor PU-1, transistor PU-2, transistor PD-1, and transistor PD-2) that form two cross-coupled inverters. The storage cell has two stable states, which are used to represent 0 and 1. Two additional access transistors (transistor PG-1 and transistor PG-2) are electrically connected to the two cross-coupled inverters and are used to control access to the memory cell during read and write operations.

FIG. 1 is a circuit diagram of a six-transistor SRAM element of some embodiments of the present disclosure. The SRAM element 10 includes a first transistor 102 formed by a pull-up transistor PU-1 and a pull-down transistor PD-1. The SRAM element 10 also includes a second transistor 104 formed by a pull-up transistor PU-2 and a pull-down transistor PD-2. In addition, the first transistor 102 and the second transistor 104 are both coupled between a power line Vdd and a power line Vss. In some embodiments, the power line Vss may be a ground potential. In some embodiments, the pull-up transistors PU-1 and PU-2 may be p-type transistors, and the pull-down transistors PD-1 and PD-2 may be n-type transistors, and the protection scope of the present disclosure is not limited thereto.

In FIG. 1, the first inverter 102 and the second inverter 104 are cross-coupled. That is, the first inverter 102 has an input connected to the output of the second inverter 104. Similarly, the second inverter 104 has an input connected to the output of the first inverter 102. The output of the first inverter 102 is called the storage node 103. Similarly, the output of the second inverter 104 is called the storage node 105. In the normal operation mode, the storage node 103 is in an opposite logical state to the storage node 105. By adopting a double cross-coupled inverter, the SRAM element 10 can use a latching structure to retain data so that as long as it is powered by Vdd, the stored data will not be lost without applying a refresh cycle.

In an SRAM element using a 6TSRAM element, cells are arranged in rows and columns. The columns of the SRAM array are formed by a bit line (signal line) pair, namely a first bit line BL and a second bit line BLB. The cells of the SRAM element are arranged between each bit line pair. As shown in FIG. 1, the SRAM element 10 is located between the bit line BL and the bit line BLB.

In FIG. 1 , the SRAM element 10 further includes a first pass-gate transistor PG-1 connected between the bit line BL and the output terminal 103 of the first inverter 102. The SRAM element 10 further includes a second pass-gate transistor PG-2 connected between the bit line BLB and the output terminal 105 of the second inverter 104. The gates of the first pass-gate transistor PG-1 and the second pass-gate transistor PG-2 are connected to the word line WL, which connects the SRAM elements in the same row in the SRAM array.

In operation, if gate transistors PG-1 and PG-2 are not turned on, the SRAM element 10 will hold the complementary values at the storage nodes 103 and 105 indefinitely as long as power is supplied by Vdd. This is because each inverter in the pair of cross-coupled inverters drives the input of the other inverter, thereby maintaining the voltage at the storage node. This condition will remain stable until the SRAM is powered off, or a write cycle is performed to change the stored data on the storage node.

In the circuit diagram of FIG. 1, the pull-up transistors PU-1 and PU-2 are p-type transistors. The pull-down transistors PD-1 and PD-2 and the gate transistors PG-1 and PG-2 are n-type transistors. However, in some other embodiments, the pull-up transistors PU-1 and PU-2 are n-type transistors, and the pull-down transistors PD-1 and PD-2 and the gate transistors PG-1 and PG-2 are p-type transistors.

The structure of the SRAM element 10 in FIG. 1 is described in the context of a 6T-SRAM. However, it should be understood by a person skilled in the art that the features of the various embodiments described herein may be used to form other types of elements, such as an 8T-SRAM memory element, or memory elements other than SRAM, such as standard cells, gated diodes, or ESD (electrostatic discharge) devices. In addition, the disclosed embodiments may be used as a standalone memory element or a memory element formed with other integrated circuits, etc.

FIG. 2A, FIG. 2B, and FIG. 2C are three-dimensional views of SRAM elements of some embodiments of the present disclosure. FIG. 2D and FIG. 2E are cross-sectional views of SRAM elements of some embodiments of the present disclosure. More specifically, the perspective view of FIG. 2C removes the non-conductive element in the perspective view of FIG. 2A. FIG. 2D and FIG. 2E are cross-sectional views along the A-A line and the B-B line of FIG. 2A, respectively.

The figure shows an SRAM element 10. The SRAM element 10 includes a substrate 90. The substrate 90 can be made of silicon or other semiconductor materials. Alternatively or additionally, the substrate 90 can include other elemental semiconductor materials such as germanium. In some embodiments, the substrate 90 is made of a compound semiconductor such as silicon carbide (SiC), gallium arsenide (GaAs), indium arsenide (InAs), or indium phosphide (InP). In some embodiments, the substrate 90 is made of an alloy semiconductor, such as silicon germanium (SiGe), silicon germanium carbide (SiGeC), gallium arsenic phosphide (GaAsP), or gallium indium phosphide (GaInP). In some embodiments, the substrate 90 includes an epitaxial layer. For example, the substrate 90 has an epitaxial layer covering a bulk semiconductor.

In some embodiments, the SRAM element 10 is a 6T-SRAM, wherein the SRAM cell of the SRAM element 10 includes a gate transistor PG-1, a gate transistor PG-2, a pull-down transistor PD-1, a pull-down transistor PD-2, a pull-up transistor PU-1, and a pull-up transistor PU-2 disposed above a substrate 90. In some embodiments, transistor PD-1, transistor PD-2, and transistor PG-1, transistor PG-2 are located at a first level LV1. On the other hand, transistor PU-1 and transistor PU-2 are located at a second level LV2 vertically above the first level LV1. That is, transistor PU-1 and transistor PU-2 are stacked on transistor PD-1, transistor PD-2, transistor PG-1, and transistor PG-2.

In more detail, as shown in the figure. In Figure 2B and Figure 2C, transistor PG-1 and transistor PD-1 are arranged along the Y direction, and transistor PG-2 and transistor PD-2 are arranged along the Y direction. Transistor PG-1 and transistor PD-2 are arranged along the X direction, and transistor PG-2 and transistor PD-1 are arranged along the X direction. As shown in the figure. In Figure 2D and Figure 2E, transistor PU-1 and transistor PD-1 are arranged along the Z direction, and transistor PU-2 and transistor PD-2 are arranged along the Z direction. In Figure 2D, transistor PU-1 overlaps vertically with transistor PD-1 along the Z direction. In Figure 2E, transistor PU-2 overlaps vertically with transistor PD-2 along the Z direction.

In different embodiments, transistor PG-1, transistor PG-2, transistor PD-1, transistor PD-2, transistor PU-1, and transistor PU-2 have a gate all around (GAA) configuration. That is, the channel region of each of transistor PG-1, transistor PG-2, transistor PD-1, transistor PD-2, transistor PU-1, and transistor PU-2 may include a plurality of semiconductor channel layers arranged in layers, and each semiconductor channel layer is surrounded by a corresponding gate structure.

Referring to FIG. 2D and FIG. 2E, transistor PG-1 may include semiconductor channel layers 112 arranged layer by layer along the Z direction. Transistor PD-1 may include semiconductor channel layers 114 arranged layer by layer along the Z direction. Transistor PD-2 may include semiconductor channel layers 116 arranged layer by layer along the Z direction. Transistor PG-2 may include semiconductor channel layers 118 arranged layer by layer along the Z direction. Transistor PU-1 may include semiconductor channel layers 212 arranged layer by layer along the Z direction. Transistor PU-2 may include semiconductor channel layers 214 arranged layer by layer along the Z direction.

Each of transistor PG-1, transistor PG-2, transistor PD-1, transistor PD-2, transistor PU-1, and transistor PU-2 of the SRAM element 10 includes a gate structure. For example, transistor PG-1 includes a gate structure 152 surrounding each semiconductor channel layer 112. Transistor PD-1 includes a gate structure 154 surrounding each semiconductor channel layer 114. Transistor PD-2 includes a gate structure 156 surrounding each semiconductor channel layer 116. Transistor PG-2 includes a gate structure 158 surrounding each semiconductor channel layer 118. Transistor PU-1 includes a gate structure 252 surrounding each semiconductor channel layer 212. Transistor PU-2 includes a gate structure 254 surrounding each semiconductor channel layer 214.

As shown in the cross-sectional views of FIG. 2D and FIG. 2E , each gate structure 152, 154, 156, 158, 252, and 254 includes a gate dielectric layer 300, a work function metal layer 302 above the gate dielectric layer 300, and a filling metal 304 above the work function metal layer 302. In some embodiments, the gate dielectric layer 300 includes a layer of high-k dielectric. In some other embodiments, the gate dielectric layer 300 includes a multi-layer structure, such as an interface layer and a high-k dielectric material. Examples of high-k dielectric materials include _HfO2 , HfSiO, HfSiON, HfTaO, HfTiO, HfZrO, zirconia, alumina, _titania , _HfO2 - _Al2O3 alloy, other suitable high-k dielectric materials, and/or combinations thereof. Examples of interface layers include silicon oxide ( _SiO2 ), silicon nitride (SiN), silicon oxynitride (SiON), hBN, _aluminum oxide ( _Al2O3 ), other suitable dielectric materials, and/or combinations thereof.

The work function metal layer 302 may be an n-type or p-type work function layer. Exemplary p-type work function metals include TiN, TaN, Ru, Mo, Al, WN, ZrSi ₂ , MoSi ₂ , TaSi ₂ , NiSi ₂ , WN, other suitable p-type work function materials, or combinations thereof. Exemplary n-type work function metals include Ti, Ag, TaAl, TaAlC, TiAlN, TaC, TaCN, TaSiN, Mn, Zr, other suitable n-type work function materials, or combinations thereof. The work function layer may include multiple layers. The fill metal 304 may include tungsten (W), aluminum (Al), copper (Cu), or other suitable conductive materials.

The SRAM element 10 also includes a gate spacer 125 disposed on opposite sidewalls of each gate structure 152, 154, 156, and 158, and a gate spacer 225 disposed on opposite sidewalls of each gate structure 252 and 254. In some embodiments, the gate spacers 125 and 225 may be formed of an insulating dielectric material, such as a silicon nitride-based material. Examples of silicon nitride-based materials may be SiN, SiON, SiOCN, or SiCN, and combinations thereof.

The SRAM element 10 further includes an inner spacer 126 vertically located between two adjacent semiconductor channel layers 112, 114, 116, and 118, and an inner spacer 226 vertically located between two adjacent semiconductor channel layers 212 and 214. In some embodiments, the inner spacers 126 and 226 may be formed of an insulating dielectric material, such as a silicon nitride-based material. Examples of silicon nitride-based materials may be SiN, SiON, SiOCN, or SiCN, and combinations thereof.

In different embodiments, each of the transistors PG-1, PG-2, PD-1, PD-2, PU-1, and PU-2 of the SRAM element 10 includes a source/drain region located on opposite sides of its respective semiconductor channel layer and located on opposite sides of its respective gate structure. For example, the transistor PG-1 includes source/drain epitaxial structures SD11 and SD12, which are located on opposite sides of the gate structure 152 and contact the two ends of each semiconductor channel layer 112. The transistor PD-1 includes source/drain epitaxial structures SD12 and SD13, which are located on opposite sides of the gate structure 154 and contact the two ends of each semiconductor channel layer 114. Transistor PD-2 includes source/drain epitaxial structures SD14 and SD15, which are respectively located on both sides of the gate structure 156 and contact the two ends of each semiconductor channel layer 116. Transistor PG-2 includes source/drain epitaxial structures SD15 and SD16, which are respectively located on both sides of the gate structure 158 and contact the two ends of each semiconductor channel layer 118. Transistor PU-1 includes source/drain epitaxial structures SD21 and SD22, which are respectively located on both sides of the gate structure 252 and contact the two ends of each semiconductor channel layer 212. Transistor PU-2 includes source/drain epitaxial structures SD23 and SD24, which are respectively located on both sides of the gate structure 254 and contact the two ends of each semiconductor channel layer 214. As shown in Figures 2D and 2E, transistor PG-1 and transistor PD-1 can share the same source/drain epitaxial structure SD12, and transistor PG-2 and transistor PD-2 can share the same source/drain epitaxial structure SD15.

The semiconductor channel layer 112, the gate structure 152, and the source/drain epitaxial structures SD11 and SD12 may collectively serve as a transistor PG-1. The semiconductor channel layer 114, the gate structure 154, and the source/drain epitaxial structures SD12 and SD13 may collectively serve as a transistor PD-1. The semiconductor channel layer 116, the gate structure 156, and the source/drain epitaxial structures SD14 and SD15 may collectively serve as a transistor PD-2. The semiconductor channel layer 118, the gate structure 158, and the source/drain epitaxial structures SD15 and SD16 may collectively serve as a transistor PG-2. The semiconductor channel layer 212, the gate structure 252, and the source/drain epitaxial structures SD21 and SD22 can be used together as transistor PU-1. The semiconductor channel layer 214, the gate structure 254, and the source/drain epitaxial structures SD23 and SD24 can be used together as transistor PU-2.

In some embodiments, the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16 may include a different conductivity type from the source/drain epitaxial structures SD21, SD22, SD23, and SD24. In some embodiments where transistor PD-1, transistor PD-2, transistor PG-1, and transistor PG-2 are n-type transistors and transistor PU-1 and transistor PU-2 are p-type transistors, the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16 are n-type epitaxial structures, and the source/drain epitaxial structures SD21, SD22, SD23, and SD24 are p-type epitaxial structures. On the other hand, in some embodiments where transistor PD-1, transistor PD-2, transistor PG-1, and transistor PG-2 are p-type transistors and transistor PU-1 and transistor PU-2 are n-type transistors, source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16 are p-type epitaxial structures, and source/drain epitaxial structures SD21, SD22, SD23, and SD24 are n-type epitaxial structures. Examples of n-type dopants may be phosphorus (P), arsenic (As), or antimony (Sb), etc. Examples of p-type dopants may be boron (B), gallium (Ga), indium (In), aluminum (Al), etc. In some embodiments, the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, SD16, SD21, SD22, SD23, and SD24 may include Si, SiGe, Ge, III-V group materials, etc. In some embodiments, the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, SD16, SD21, SD22, SD23, and SD24 may include epitaxial materials for N-type devices (e.g., NFETs), such as SiP, SiAs, silicon carbide, etc. On the other hand, the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, SD16, SD21, SD22, SD23 and SD24 may include epitaxial materials for P-type devices (eg, PFETs), such as SiGeB, SiCB, GeSn (eg, Ge _0.9 Sn _0.1 ), and the like.

The SRAM element 10 also includes source/drain contacts 171, 172, 173, 174, 175, 176, 177, 176, 271, 272, 273, and 274, covering source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, SD16, SD21, SD22, SD23, and SD24, respectively. In some embodiments, the source/drain contacts 171, 172, 173, 174, 175, 176, 177, 176, 271, 272, 273, and 274 may be made of a conductive material such as metal. The conductive material may include one or more layers of Co, Ni, W, Ti, Ta, Cu, Al, TiN, and TaN, or any other suitable material.

As shown in FIG. 2B and FIG. 2C, the SRAM element 10 further includes metal lines ML11, ML12, ML13, and ML14. In some embodiments, the metal lines ML11, ML12, ML13, and ML14 are disposed above the substrate 90 and arranged along the X direction. In some embodiments, the metal lines ML11, ML12, ML13, and ML14 each include a lengthwise direction extending along the Y direction. The horizontal position of the metal lines ML11, ML12, ML13, and ML14 is located below the transistor PD-1, the transistor PD-2, the transistor PG-1, and the transistor PG-2.

In some embodiments, each of the metal lines ML11, ML12, ML13, and ML14 has a width (along the X direction) in the range of about 1 nm to about 1 μm, a length (along the Y direction) in the range of about 100 nm to about 20 μm, and a thickness (along the Z direction) in the range of about 1 nm to about 1 μm.

The SRAM element 10 further includes metal vias MV11, MV12, MV13, and MV14. In some embodiments, when viewed from above (see FIG. 13), the metal via MV11 is at least partially aligned with the metal via MV14 along the X direction. The metal via MV12 is at least partially aligned with the metal via MV13 along the X direction.

In some embodiments, the metal via MV11 contacts the bottom surface of the source/drain contact 173 covering the source/drain epitaxial structure SD13 and the top surface of the metal line ML11. That is, the metal line ML11 is electrically connected to the source/drain epitaxial structure SD13 of the transistor PD-1.

In some embodiments, the metal via MV12 contacts the bottom surface of the source/drain contact 171 covering the source/drain epitaxial structure SD11 and the top surface of the metal line ML12. That is, the metal line ML12 is electrically connected to the source/drain epitaxial structure SD11 of the transistor PG-1.

In some embodiments, the metal via MV13 contacts the bottom surface of the source/drain contact 174 covering the source/drain epitaxial structure SD14 and the top surface of the metal line ML13. That is, the metal line ML13 is electrically connected to the source/drain epitaxial structure SD14 of the transistor PD-2.

In some embodiments, the metal via MV14 contacts the bottom surface of the source/drain contact 176 covering the source/drain epitaxial structure SD16 and the top surface of the metal line ML14. That is, the metal line ML14 is electrically connected to the source/drain epitaxial structure SD16 of the transistor PG-2.

In some embodiments where transistors PU-1 and PU-2 are p-type transistors and transistors PD-1, PD-2, PG-1, and PG-2 are n-type transistors, metal lines ML11 and ML13 may be used together as the power line Vss of the SRAM element 10 (see FIG. 1 ). In other words, metal lines ML11 and ML13 may be considered as the first and second parts of the power line Vss. Metal line ML12 may be used as the bit line BL of the SRAM element 10 (see FIG. 1 ), and metal line ML14 may be used as the bit line BLB of the SRAM element 10 (see FIG. 1 ). Alternatively, metal line ML12 may be used as the bit line BLB of the SRAM element 10, and metal line ML14 may be used as the bit line BL of the SRAM element 10.

In some embodiments where transistors PU-1 and PU-2 are n-type transistors and transistors PD-1, PD-2, PG-1, and PG-2 are p-type transistors, metal lines ML11 and ML13 may be used together as a power line Vdd of the SRAM element 10. In other words, metal lines ML11 and ML13 may be considered as the first and second parts of the power line Vdd. Metal line ML12 may be used as a bit line BL of the SRAM element 10, and metal line ML14 may be used as a bit line BLB of the SRAM element 10. Alternatively, metal line ML12 may be used as a bit line BLB of the SRAM element 10, and metal line ML14 may be used as a bit line BL of the SRAM element 10.

The SRAM element 10 further includes metal vias MV21, MV22, and MV23. In some embodiments, the horizontal positions of the metal vias MV21, MV22, and MV23 are located between the first level LV1 perpendicular to the transistors PD-1, PD-2, PG-1, and PG-2 and the second level LV2 of the transistors PU-1 and PU-2. In some embodiments, the metal via MV21 contacts the top surface of the gate structure 154 and the bottom surface of the gate structure 252. That is, the gate structure 154 of the transistor PD-1 is electrically connected to the gate structure 252 of the transistor PU-1.

In some embodiments, the metal via MV22 contacts the top surface of the source/drain contact 172 covering the source/drain epitaxial structure SD12 and the bottom surface of the source/drain contact 271 covering the source/drain epitaxial structure SD21. That is, the source/drain epitaxial structure SD12 of the transistor PD-1 and the transistor PG-1 is electrically connected to the source/drain epitaxial structure SD21 of the transistor PU-1.

In some embodiments, the metal via MV23 contacts the top surface of the source/drain contact 175 covering the source/drain epitaxial structure SD15 and the bottom surface of the source/drain contact 274 covering the source/drain epitaxial structure SD24. That is, the source/drain epitaxial structure SD15 of the transistor PD-2 and the transistor PG-2 is electrically connected to the source/drain epitaxial structure SD24 of the transistor PU-2.

In some embodiments, there is also a metal via MV24 (see FIG. 19, not shown in FIG. 2B and FIG. 2C because it is shielded by metal vias MV21, MV22, and MV23) that contacts the top surface of gate structure 156 and the bottom surface of gate structure 254. That is, gate structure 156 of transistor PD-2 is electrically connected to gate structure 254 of transistor PU-2.

The SRAM element 10 further includes metal vias MV31, MV32, MV33, and MV34, and metal lines ML31 and ML32. In some embodiments, the horizontal positions of the metal vias MV31, MV32, MV33, and MV34 and the metal lines ML31 and ML32 are vertically higher than the second level LV2 of the transistors PU-1 and PU-2. In some embodiments, the metal via MV31 contacts the top surface of the source/drain contact 271 covering the source/drain epitaxial structure SD21 and the bottom surface of the metal line ML31. The metal via MV32 contacts the top surface of the gate structure 254 and the bottom surface of the metal line ML31. Therefore, the source/drain epitaxial structure SD21 of transistor PU-1 is electrically connected to the gate structure 254 of transistor PU-2.

In some embodiments, metal via MV33 contacts the top surface of gate structure 252 and the bottom surface of metal line ML32. Metal via MV34 contacts the top surface of source/drain contact 274 covering source/drain epitaxial structure SD24 and the bottom surface of metal line ML32. Therefore, source/drain epitaxial structure SD24 of transistor PU-2 is electrically connected to gate structure 252 of transistor PU-1.

The SRAM element 10 also includes a metal wire ML41 and metal vias MV41 and MV42. In some embodiments, the horizontal position of the metal wire ML41 is vertically higher than the second level LV2 of the transistor PU-1, the transistor PU-2 and higher than the horizontal position of the metal wires ML31 and ML32. In some embodiments, the metal via MV41 contacts the top surface of the gate structure 152 and the bottom surface of the metal wire ML41. Therefore, the gate structure 152 of the transistor PG-1 is electrically connected to the metal wire ML41. In some embodiments, the metal via MV42 contacts the top surface of the gate structure 158 and the bottom surface of the metal wire ML41. Therefore, the gate structure 158 of the transistor PG-2 is electrically connected to the metal wire ML41.

In some embodiments, the metal line ML41 can serve as the word line WL of the SRAM element 10 (see FIG. 1 ). Therefore, the gate structure 152 of the transistor PG-1 and the gate structure 158 of the transistor PG-1 are electrically connected to the word line WL.

The SRAM element 10 also includes metal lines ML51 and ML52, and metal vias MV51 and MV52. In some embodiments, the horizontal position of the metal lines ML51 and ML52 is vertically higher than the second level LV2 of the transistor PU-1, the transistor PU-2, and higher than the horizontal position of the metal lines ML41 and ML42. In some embodiments, the metal via MV51 contacts the top surface of the source/drain contact 272 covering the source/drain epitaxial structure SD22 and the bottom surface of the metal line ML51. Therefore, the source/drain epitaxial structure SD22 of the transistor PU-1 is electrically connected to the metal line ML51.

In some embodiments, the metal via MV52 contacts the top surface of the source/drain contact 273 covering the source/drain epitaxial structure SD23 and the bottom surface of the metal line ML52. Therefore, the source/drain epitaxial structure SD23 of the transistor PU-2 is electrically connected to the metal line ML52.

In some embodiments where transistors PU-1 and PU-2 are p-type transistors and transistors PD-1, PD-2, PG-1, and PG-2 are n-type transistors, metal lines ML51 and ML52 can serve together as the power line Vdd of the SRAM element 10 (see FIG. 1 ). In other words, metal lines ML51 and ML52 can be regarded as the first and second parts of the power line Vdd.

In some embodiments where transistors PU-1 and PU-2 are n-type transistors and transistors PD-1, PD-2, PG-1, and PG-2 are p-type transistors, metal lines ML51 and ML52 may serve together as a power line Vss of the SRAM element 10. In other words, metal lines ML51 and ML52 may be considered as the first and second parts of the power line Vss.

The SRAM element 10 also includes semiconductor strips 312 and 314 protruding above the substrate 90. In some embodiments, the semiconductor strips 312 and 314 may include the same material as the substrate 90. For example, the semiconductor strips 312 and 314 may be made of silicon (Si) or other suitable semiconductor materials. In some embodiments, each semiconductor strip 312 and 314 extends along the Y direction. Transistor PG-1 and transistor PD-1 are disposed above the semiconductor strip 312, and transistor PG-2 and transistor PD-2 are disposed above the semiconductor strip 314.

In some embodiments, the top surfaces of the metal lines ML11, ML12, ML13, and ML14 are lower than the top surfaces of the semiconductor strips 312 and 314. The metal lines ML11 and ML12 are disposed on opposite sides of the semiconductor strip 312, and the metal lines ML13 and ML14 are disposed on opposite sides of the semiconductor strip 314. In some embodiments, the metal lines ML12 and ML13 are located between the semiconductor strips 312 and 314.

As shown in FIG. 2C, along the X direction, the distance D1 between metal lines ML12 and ML13 is smaller than the distance D2 between metal lines ML11 and ML12, and smaller than the distance D3 between metal lines ML13 and ML14. This is because there is no semiconductor strip between metal lines ML12 and ML13 (see FIG. 2A).

Referring to the metal line ML11 and the metal via MV11 (see FIG. 2A ), the SRAM element 10 further includes a liner 321 located on the sidewall of the metal line ML11 and on the opposite side of the metal via MV11. In some embodiments, the liner 321 may further extend to the top surface of the metal line ML11. The SRAM element 10 further includes a dielectric material 331 in contact with the sidewall of the metal via MV11. In some embodiments, the dielectric material 331 is separated from the metal line ML11 by the liner 321.

Referring to the metal line ML12 and the metal via MV12 (see FIGS. 2A and 2B ), the SRAM element 10 further includes a liner 322 located on the sidewall of the metal line ML12 and on the opposite side of the metal via MV12 (not shown in FIG. 2A due to being shielded). In some embodiments, the liner 322 may further extend to the top surface of the metal line ML12. The SRAM element 10 further includes a dielectric material 332 in contact with the sidewall of the metal via MV12 (not shown in FIG. 2A due to being shielded). In some embodiments, the dielectric material 332 is separated from the metal line ML12 by the liner 322.

Referring to the metal line ML13 and the metal via MV13 (see FIGS. 2A and 2B ), the SRAM element 10 further includes a liner 323 located on the sidewall of the metal line ML13 and on the opposite side of the metal via MV13 (not shown in FIG. 2A due to being shielded). In some embodiments, the liner 323 may further extend to the top surface of the metal line ML13. The SRAM element 10 further includes a dielectric material 333 in contact with the sidewall of the metal via MV13 (not shown in FIG. 2A due to being shielded). In some embodiments, the dielectric material 333 is separated from the metal line ML13 by the liner 323.

Referring to the metal line ML14 and the metal via MV14 (see FIG. 2A and FIG. 2B ), the SRAM element 10 further includes a liner 324 located on the sidewall of the metal line ML14 and on the opposite side of the metal via MV14. In some embodiments, the liner 324 may further extend to the top surface of the metal line ML14. The SRAM element 10 further includes a dielectric material 334 in contact with the sidewall of the metal via MV14. In some embodiments, the dielectric material 334 is separated from the metal line ML14 by the liner 324.

The SRAM cell 10 further includes isolation structures 315 and 316 above the substrate 90 and adjacent to the semiconductor strips 312 and 314. In more detail, the isolation structure 315 is adjacent to the semiconductor strip 312, wherein the metal line ML11 is located between the semiconductor strip 312 and the isolation structure 315. In some embodiments, the liner 321 also contacts the semiconductor strip 312, the isolation structure 315, and the top surface of the substrate 90. Similarly, the isolation structure 316 is adjacent to the semiconductor strip 314, wherein the metal line ML14 is between the semiconductor strip 314 and the isolation structure 316. In some embodiments, liner 324 also contacts semiconductor strip 314, isolation structure 316, and the top surface of substrate 90.

In some embodiments, liner 322 contacts semiconductor strip 312, liner 323, and the top surface of substrate 90. Liner 323 contacts semiconductor strip 314, liner 322, and the top surface of substrate 90.

Liners 321, 322, 323, and 324 may be made of a suitable dielectric material. For example, liner layers 321, 322, 323, and 324 may be made of silicon nitride. Dielectric materials 331, 332, 333, and 334 may be made of a suitable dielectric material. For example, dielectric materials 331, 332, 333, and 334 may be made of silicon oxide. Isolation structures 315 and 316 may be shallow trench isolation (STI) structures, suitable isolation structures, combinations thereof, and the like. In some embodiments, isolation structures 315 and 316 may be made of oxides (e.g., silicon oxide), nitrides (e.g., silicon nitride), or combinations thereof. In some embodiments, isolation structures 315 and 316 are made of the same material as dielectric materials 331, 332, 333, and 334.

The SRAM element also includes an interlayer dielectric layer 150 that laterally surrounds transistor PD-1, transistor PG-1, transistor PD-2, and transistor PG-2. The SRAM element also includes a semiconductor layer 180 disposed above the interlayer dielectric layer 150. The SRAM element also includes an interlayer dielectric layer 185 disposed above the semiconductor layer 180 and laterally surrounding transistor PU-1 and transistor PU-2. The SRAM element also includes intermetallic dielectric layers 191, 192, 193, 194, 195, and 196.

In some embodiments, metal vias MV21, MV22, MV23, and MV24 are disposed in semiconductor layer 180. Metal vias MV31, MV32, MV33, and MV34 are disposed in intermetallic dielectric layer 191. Metal lines ML31 and ML32 are disposed in intermetallic dielectric layer 192. Metal line ML41 is disposed in intermetallic dielectric layer 194. Metal lines ML51 and ML52 are disposed in intermetallic dielectric layer 196. Metal vias MV41 and MV42 are disposed in semiconductor layer 180 and intermetallic dielectric layers 191, 192, 193, and 194. Metal vias MV51 and MV52 are provided in intermetallic dielectric layers 191, 192, 193, 194 and 195.

In some embodiments, interlayer dielectric layers 150 and 185 and intermetallic dielectric layers 191, 192, 193, 194, 195, and 196 may include silicon oxide, silicon nitride, silicon oxynitride, tetraethoxysilane (TEOS), phosphosilicate glass (PSG), borophosphosilicate glass (BPSG), low-k dielectric materials, and/or other suitable dielectric materials. Examples of low-k dielectric materials include, but are not limited to, fluorinated quartz glass (FSG), carbon-doped silicon oxide, amorphous fluorinated carbon, polyparaxylene, bisbenzocyclobutene (BCB), or polyimide.

In some embodiments, the metal wires and metal vias may be made of suitable conductive materials, such as metal. In some embodiments, the conductive materials may include Pt, Ti, TiN, Al, W, WN, Ru, RuO, Ta, Ni, Co, Cu, Ag, Au, etc.

Figures 3 to 31 show multiple stages of a method for forming an SRAM element in some embodiments of the present disclosure.

Refer to FIG. 3. A substrate 90 is provided. Then, semiconductor layers 110 and semiconductor layers 111 are alternately deposited on the substrate 90. In some embodiments, the semiconductor layer 110 is made of SiGe. For example, the germanium percentage (atomic percentage concentration) of the semiconductor layer 110 ranges between about 10% and about 50%, and may have a higher or lower germanium percentage. However, it should be understood that the values listed throughout the discussion are merely examples and may be changed to different values. For example, the semiconductor layer 110 may vary from Si _0.5 Ge _0.5 to Si _0.9 Ge _0.1 , where the ratio between Si and Ge may vary depending on the embodiment, and the present disclosure is not limited thereto. In some embodiments, semiconductor layer 111 may be a pure silicon layer that does not contain germanium. Semiconductor layer 111 may also be a substantially pure silicon layer, for example, having a germanium percentage of less than about 1%. In some embodiments, semiconductor layers 110 and 111 may be deposited on substrate 90 using a suitable deposition process, such as chemical vapor deposition (CVD), molecular beam epitaxy (MBE), or other suitable processes.

Referring to FIG. 4A, FIG. 4B, and FIG. 4C, FIG. 4B and FIG. 4C are cross-sectional views along line B-B and line C-C of FIG. 4A, respectively. The semiconductor layers 110 and 111 and the substrate 90 are patterned to form fin structures FN11 and FN12. In some embodiments, the fin structure FN11 includes a semiconductor strip 312 and a stack of semiconductor layers 110 and 111 formed above the semiconductor strip 312. The fin structure FN12 includes a semiconductor strip 314 and a stack of semiconductor layers 110 and 111 formed above the semiconductor strip 314. In some embodiments, the semiconductor layers 110 and 111 and the substrate 90 may be patterned by, for example, forming a patterned mask over the stack of semiconductor layers 110 and 111, which defines the pattern of the fin structures FN11 and FN12, and then performing an etching process to remove unnecessary portions of the semiconductor layers 110 and 111 and the substrate 90. After the patterning process, trenches are formed in the substrate 90, and semiconductor strips 312 and 314 are formed between two adjacent trenches.

Refer to FIG. 5. Isolation structures 315, 316, and 317 are formed on substrate 90 and laterally surround fin structures FN11 and FN12. More specifically, fin structure FN11 is located between isolation structures 315 and 317, and fin structure FN12 is located between isolation structures 316 and 317. Isolation structures 315, 316, and 317 can be formed by, for example, depositing a dielectric material on substrate 90 and filling the area outside fin structures FN11 and FN12, and then performing a planarization process (e.g., chemical mechanical polishing; CMP) to remove excess dielectric material until the topmost semiconductor layer 110 is exposed.

Then, a patterned mask MA1 is formed over the substrate 90. The patterned mask MA1 includes an opening O1 that exposes a portion of the isolation structures 316 and 317. In some embodiments, the fin structures FN11, FN12, and the isolation structure 315 are covered by the patterned mask MA1. In some embodiments, the patterned mask MA1 may be a photoresist.

Referring to FIG. 6 , an etching process is performed to remove the exposed portions of the isolation structures 316 and 317 to form the trench TR1. After the trench TR1 is formed, the patterned mask MA1 is removed.

Refer to FIG. 7. Metal lines ML12 and ML14 are formed in the trench TR1, and liner layers 322 and 324 are formed on the metal lines ML12 and ML14, respectively. In some embodiments, a liner material (e.g., a dielectric material) is deposited along the trench TR1, and then a conductive material is deposited on the liner material, and a planarization process (e.g., CMP) is performed to remove excess liner material and excess conductive material until the topmost semiconductor layer 110 is exposed. Thus, metal lines ML12 and ML14 and liner layers 322 and 324 are formed.

Refer to FIG. 8. A patterned mask MA2 is formed over the substrate 90. The patterned mask MA2 includes an opening O2 that exposes a portion of the isolation structures 315 and 317. In some embodiments, the fin structures FN1, FN2, the isolation structure 316, the metal lines ML12 and ML14, and the liner layers 322 and 324 are covered by the patterned mask MA2. In some embodiments, the patterned mask MA2 may be a photoresist.

Referring to FIG. 9 , an etching process is performed to remove the exposed portions of the isolation structures 315 and 317 to form the trench TR2. After the trench TR2 is formed, the patterned mask MA2 is removed. In some embodiments, after the etching process is completed, the isolation structure 315 between the fin structures FN11 and FN12 is completely removed.

Refer to FIG. 10. Metal lines ML11 and ML13 are formed in the trench TR2, and liner layers 321 and 323 are formed on the metal lines ML11 and ML13, respectively. In some embodiments, a liner material (e.g., a dielectric material) is deposited along the trench TR2, and then a conductive material is deposited on the liner material, and a planarization process (e.g., CMP) is performed to remove excess liner material and excess conductive material until the topmost semiconductor layer 110 is exposed. Thus, metal lines ML11 and ML13 and liner layers 321 and 323 are formed. Metal lines ML12 and ML14 are formed before metal lines ML11 and ML13. However, in other embodiments, metal lines ML11 and ML13 may be formed before metal lines ML12 and ML14.

Refer to FIG. 11. The metal lines ML11, ML12, ML13, and ML14 are etched back using the fin structures FN11 and FN12, the isolation structures 315 and 316, and the liner layers 321, 322, 323, and 324 as etching masks. After the etching back process, the top surfaces of the metal lines ML11, ML12, ML13, and ML14 are lowered to be lower than the top surfaces of the semiconductor strips 312 and 314.

After the metal lines ML11, ML12, ML13, and ML14 are etched back, a liner material is formed to cover the top surface of the etched back metal lines ML11, ML12, ML13, and ML14. In some embodiments, the liner material may be the same as the material of the liner layers 321, 322, 323, and 324. Therefore, the liner material on the etched back metal lines ML11, ML12, ML13, and ML14 may be regarded as an extension of the liner layers 321, 322, 323, and 324, respectively.

Next, dielectric materials 331, 332, 333, and 334 are formed on metal lines ML11, ML12, ML13, and ML14, respectively. In some embodiments, dielectric materials 331, 332, 333, and 334 can be formed by, for example, depositing dielectric materials on substrate 90 and filling spaces in liner layers 321, 322, 323, and 324, and then performing a planarization process (e.g., CMP) to remove excess dielectric materials until the topmost semiconductor layer 110 is exposed.

Referring to FIG. 12 , a patterned mask MA3 is formed over a substrate 90 . The patterned mask MA3 includes an opening O3 that exposes a portion of the dielectric materials 331 , 332 , 333 , and 334 . Then, an etching process is performed to remove the exposed portions of the dielectric materials 331 , 332 , 333 , and 334 through the opening O3 of the patterned mask MA3 , and to form trenches TR3 in the dielectric materials 331 , 332 , 333 , and 334 until the metal lines ML11 , ML12 , ML13 , and ML14 are exposed. In some embodiments, the extended portions of the liner layers 321 , 322 , 323 , and 324 may be etched to expose the metal lines ML11 , ML12 , ML13 , and ML14 thereunder. In some embodiments, the trench TR3 defines the locations of metal vias MV11, MV12, MV13, and MV14 formed in subsequent steps.

Refer to FIG. 13. After forming the trench TR3, the patterned mask MA3 is removed. Then, metal vias MV11, MV12, MV13, and MV14 are formed in the trench TR3 and contact the metal lines ML11, ML12, ML13, and ML14, respectively. In some embodiments, the metal vias MV11, MV12, MV13, and MV14 can be formed by, for example, filling the trench TR3 with a conductive material and performing a planarization process (e.g., CMP) to remove excess conductive material until the topmost semiconductor layer 110 is exposed.

Refer to FIG. 14. Using the fin structures FN11 and FN12 as etching masks, metal vias MV11, MV12, MV13 and MV14, isolation structures 315 and 316, liner layers 321, 322, 323 and 324, and dielectric materials 331, 332, 333 and 334 are etched back. In some embodiments, the etching back process is performed to lower the positions of the metal vias MV11, MV12, MV13 and MV14, isolation structures 315 and 316, liner layers 321, 322, 323 and 324, and dielectric materials 331, 332, 333 and 334 to below the bottommost semiconductor layer 111.

Referring to FIG. 15A, FIG. 15B, and FIG. 15C, FIG. 15B and FIG. 15C are cross-sectional views along line B-B and line C-C of FIG. 15A, respectively. Dummy gate structures 132, 134, 136, and 138 are formed over substrate 90 and span fin structures FN11 and FN12. In some embodiments, each of the dummy gate structures 132, 134, 136, and 138 includes a dummy gate dielectric 130 and a dummy gate electrode 131 over the dummy gate dielectric 130. The dummy gate dielectric 130 may be, for example, silicon oxide, silicon nitride, a combination thereof, and the like, and may be formed by deposition or thermal growth using an acceptable technique. The dummy gate electrode 131 may be a conductive or non-conductive material, and may include amorphous silicon, polysilicon, polycrystalline silicon germanium (poly-SiGe), metal nitride, metal silicide, metal oxide, and metal.

The dummy gate structures 132, 134, 136, and 138 may be formed by, for example, depositing a dummy gate dielectric layer and a dummy gate layer over the substrate 90 and then patterning the dummy gate dielectric layer and the dummy gate layer. In some embodiments, the dummy gate electrode 131 may be deposited by physical vapor deposition (PVD), chemical vapor deposition (CVD), sputter deposition, or other techniques for depositing selected materials. In some embodiments, the dummy gate dielectric 130 may be formed by thermal oxidation.

As shown in FIG. 15B and FIG. 15C, gate spacers 125 are formed on opposite sidewalls of each dummy gate structure 132, 134, 136, and 138, respectively. In some embodiments, the gate spacers 125 can be formed by, for example, depositing a spacer layer on the substrate 90, and then performing an anisotropic etching process to remove the horizontal portion of the spacer layer, so that the vertical portion of the spacer layer remains on the sidewalls of the dummy gate structures 132, 134, 136, and 138.

After forming the gate spacer 125, an etching process is performed to remove the stacked portions of the semiconductor layers 110 and 111 exposed by the dummy gate structures 132, 134, 136, and 138 and the gate spacer 125 to form source/drain grooves in the fin structures FN11 and FN12. In some embodiments, the remaining portions of the semiconductor layer 111 covered by the dummy gate structures 132, 134, 136, and 138 may be referred to as semiconductor channel layers 112, 114, 116, and 118, respectively.

Refer to FIG. 16A, FIG. 16B and FIG. 16C, wherein FIG. 16B and FIG. 16C are cross-sectional views along lines B-B and C-C of FIG. 16A, respectively. The portion of the semiconductor layer 110 exposed by the source/drain groove is laterally etched to form a sidewall groove, and then an inner spacer 126 is formed in the sidewall groove (see FIG. 16B and FIG. 16C). The inner spacer 126 can be deposited by a conformal deposition process such as CVD, ALD, etc. The inner spacer 126 can be formed, for example, by depositing an inner spacer layer on the substrate 90 and filling the sidewall groove of the semiconductor layer 110, and then performing anisotropic etching to remove the spacer layer outside the sidewall groove.

Then, source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16 are formed in the source/drain grooves, respectively. In some embodiments, the formation of the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16 may include a selective epitaxial growth (SEG) process.

Referring to FIG. 17A, FIG. 17B and FIG. 17C, FIG. 17B and FIG. 17C are cross-sectional views along the B-B line and the C-C line of FIG. 17A, respectively. An interlayer dielectric layer 150 is formed on the substrate 90 and covers the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15 and SD16. In some embodiments, the interlayer dielectric layer 150 can be formed by, for example, depositing a dielectric material on the substrate 90, and then performing a CMP process to remove excess dielectric material until the dummy gate structures 132, 134, 136 and 138 are exposed.

After the interlayer dielectric layer 150 is formed, the dummy gate structures 132, 134, 136, and 138 are removed to form gate trenches between each pair of gate spacers 125. Then, the portion of the semiconductor layer 110 exposed by the gate trenches is removed, so that the semiconductor channel layers 112, 114, 116, and 118 are suspended above the substrate 90.

Metal gate structures 152, 154, 156, and 158 are formed over the substrate 90. In more detail, the metal gate structures 152, 154, 156, and 158 are formed in gate trenches between the gate spacers 125 and surround the corresponding semiconductor channel layers 112, 114, 116, and 118. In some embodiments, the metal gate structures 152, 154, 156 and 158 can be formed by the following methods, for example, sequentially depositing a gate dielectric layer 300, a work function metal layer 302 and a filling metal 304 on the substrate 90, and filling the gate trench between the gate spacers 125, and then performing a CMP process to remove excess material of the gate dielectric layer 300, excess material of the work function metal layer 302 and excess material of the filling metal 304 until the interlayer dielectric layer 150 is exposed.

Referring to FIG. 18A, FIG. 18B and FIG. 18C, FIG. 18B and FIG. 18C are cross-sectional views along line B-B and line C-C of FIG. 18A, respectively. A portion of the interlayer dielectric layer 150 is removed to form openings exposing source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16. Then, source/drain contacts 171, 172, 173, 174, 175, and 176 are formed in the openings and cover the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16, respectively. In some embodiments, source/drain contacts 171, 172, 173, 174, 175, and 176 may be formed by, for example, depositing a conductive material in the openings and then performing a CMP process to remove excess conductive material until gate structures 152, 154, 156, and 158 are exposed.

Refer to FIG. 19. A semiconductor layer 180 is formed over the interlayer dielectric layer 150. In some embodiments, the semiconductor layer 180 may be formed over the interlayer dielectric layer 150 using a suitable deposition process. In some other embodiments, the semiconductor layer 180 may be bonded to the interlayer dielectric layer 150 using a suitable bonding process.

After forming the semiconductor layer 180 on the interlayer dielectric layer 150, metal vias MV21, MV22, MV23, MV24, MV201, and MV202 are formed in the semiconductor layer 180. In some embodiments, the metal vias MV21, MV22, MV23, MV24, MV201, and MV202 may be formed by, for example, patterning the semiconductor layer 180 to form an opening in the semiconductor layer 180, filling the opening with a conductive material, and then performing a CMP process to remove excess conductive material until the semiconductor layer 180 is exposed. In some embodiments, metal vias MV21, MV22, MV23, and MV24 contact the top surfaces of gate structure 154, source/drain contact 172, source/drain contact 175, and gate structure 156, respectively. In addition, metal vias MV201 and MV202 contact the top surfaces of gate structure 152 and gate structure 158, respectively.

Refer to Figure 20. A stack of alternating semiconductor layers 210, 211, and semiconductor layer 181 are formed above semiconductor layer 180. In some embodiments, semiconductor layer 210 is made of germanium and semiconductor layer 211 is made of germanium tin (GeSn). In some embodiments, the germanium percentage (atomic percentage concentration) of semiconductor layer 211 is approximately 95%. For example, semiconductor layer 211 may include Ge _0.95 Sn _0.05 . In some embodiments, the germanium percentage of semiconductor layer 211 is greater than the germanium percentage of semiconductor layer 210. In some embodiments, semiconductor layer 181 may be made of silicon.

In some embodiments, a stack of alternating semiconductor layers 210, 211 may be deposited on semiconductor layer 181. Thereafter, the structure of semiconductor layers 210, 211 and semiconductor layer 181 is transferred over semiconductor layer 180 and bonded to semiconductor layer 180.

In other embodiments, semiconductor layers 210 and 211 may be deposited on semiconductor layer 180 using a suitable deposition process, and then semiconductor layer 181 is formed on top of the topmost semiconductor layer 210. Semiconductor layers 210 and 211 and semiconductor layer 181 may be deposited using a suitable deposition process, such as chemical vapor deposition (CVD), molecular beam epitaxy (MBE), or other suitable processes.

Refer to FIG. 21. The semiconductor layer 181 is etched back to expose the topmost semiconductor layer 210. Then, the stack of semiconductor layer 210 and semiconductor layer 211 is patterned to form fin structures FN21 and FN22. In some embodiments, each of the fin structures FN21 and FN22 includes a stack of semiconductor layers 210 and 211. The semiconductor layer 211 within the fin structure FN21 can be referred to as a semiconductor channel layer 212, and the semiconductor layer 211 within the fin structure FN22 can be referred to as a semiconductor channel layer 214.

Referring to FIG. 22A, FIG. 22B, and FIG. 22C, FIG. 22B and FIG. 22C are cross-sectional views along line B-B and line C-C of FIG. 22A, respectively. Dummy gate structures 232 and 234 are formed over semiconductor layer 180 and across fin structures FN21 and FN22. In some embodiments, each of the dummy gate structures 232 and 234 includes a dummy gate dielectric 230 and a dummy gate electrode 231 over the dummy gate dielectric 230. The dummy gate dielectric 230 may be, for example, silicon oxide, silicon nitride, a combination thereof, and the like, and may be formed by deposition or thermal growth using an acceptable technique. The dummy gate electrode 231 may be a conductive or non-conductive material and may include amorphous silicon, polysilicon, polycrystalline silicon germanium (poly-SiGe), metal nitrides, metal silicides, metal oxides, and metals. The dummy gate structures 232 and 234 are formed using a method similar to the above-described formation of the dummy gate structures 132, 134, 136, and 138, and therefore will not be described in detail for the sake of brevity.

Referring to FIG. 22B and FIG. 22C , gate spacers 225 are formed on opposite sidewalls of each dummy gate structure 232 and 234, respectively. In some embodiments, the gate spacers 225 may be formed using a method similar to the above-described method for forming the gate spacers 125, and thus will not be described in detail for the sake of brevity.

Referring to FIG. 23A, FIG. 23B and FIG. 23C, FIG. 23B and FIG. 23C are cross-sectional views along the B-B line and the C-C line of FIG. 23A, respectively. Part of the semiconductor layer 210 is laterally etched to form a sidewall groove, and then an inner spacer 226 is formed in the sidewall groove. The inner spacer 226 can be formed using a method similar to that of forming the inner spacer 126.

Source/drain epitaxial structures SD21, SD22, SD23, and SD24 are formed on opposite sides of the dummy gate structures 232 and 234, respectively. In some embodiments, the source/drain epitaxial structures SD21, SD22, SD23, and SD24 can be formed using a method similar to the method for forming the source/drain epitaxial structures SD11, SD12, SD13, SD14, SD15, and SD16, so the relevant description is omitted for brevity.

Referring to FIG. 24A, FIG. 24B and FIG. 24C, FIG. 24B and FIG. 24C are cross-sectional views along line B-B and line C-C of FIG. 24A, respectively. An interlayer dielectric layer 185 is formed over the semiconductor layer 180 and covers the source/drain epitaxial structures SD21, SD22, SD23, and SD24. In some embodiments, the interlayer dielectric layer 185 can be formed by, for example, depositing a dielectric material on the semiconductor layer 180, followed by a CMP process to remove excess dielectric material until the dummy gate structures 232 and 234 are exposed.

After forming the interlayer dielectric layer 185, the dummy gate structures 232 and 234 are removed to form a gate trench between each pair of gate spacers 225. Then, the portion of the semiconductor layer 210 exposed through the gate trench is removed, so that the semiconductor channel layers 212 and 214 are suspended above the substrate 90.

Metal gate structures 252 and 254 are formed over the substrate 90. In more detail, the metal gate structures 252 and 254 are formed in the gate trenches between the gate spacers 225 and surround the semiconductor channel layers 212 and 214, respectively. In some embodiments, the metal gate structures 252 and 254 can be formed by the following methods, for example, sequentially depositing a gate dielectric layer 300, a work function metal layer 302, and a filling metal 304 on a substrate 90, and filling the gate trenches between the gate spacers 225, followed by a CMP process to remove excess material of the gate dielectric layer 300, excess material of the work function metal layer 302, and excess material of the filling metal 304 until the interlayer dielectric layer 185 is exposed.

Referring to FIG. 25A, FIG. 25B and FIG. 25C, FIG. 25B and FIG. 25C are cross-sectional views along line B-B and line C-C of FIG. 25A, respectively. A portion of the interlayer dielectric layer 185 is removed to form openings exposing source/drain epitaxial structures SD21, SD22, SD23 and SD24, and the openings expose metal vias MV201 and MV202 in the semiconductor layer 180. Then, source/drain contacts 271, 272, 273 and 274 are formed in the openings to cover the source/drain epitaxial structures SD21, SD22, SD23 and SD24, respectively. In addition, metal vias MV301 and MV302 are formed in the opening and contact the metal vias MV201 and MV202 in the semiconductor layer 180, respectively. In some embodiments, the source/drain contacts 271, 272, 273, and 274 and the metal vias MV301 and MV302 can be formed by, for example, depositing a conductive material in the opening and then performing a CMP process to remove excess conductive material until the gate structures 252 and 254 are exposed.

Refer to FIG. 26. An intermetallic dielectric layer 191 is formed over the interlayer dielectric layer 185. Then, metal vias MV31, MV32, MV33, MV34, MV401, MV402, MV403, and MV404 are formed in the intermetallic dielectric layer 191. In more detail, the metal vias MV31, MV32, MV33, and MV34 contact the source/drain contact 271, the gate structure 254, the gate structure 252, and the source/drain contact 274, respectively. The metal vias MV401 and MV402 contact the metal vias MV301 and MV302, respectively. Metal vias MV403 and MV404 contact source/drain contacts 272 and 273 respectively.

Refer to FIG. 27. An intermetallic dielectric layer 192 is formed over the intermetallic dielectric layer 191. Then, metal lines ML31 and ML32 and metal vias MV501, MV502, MV503, and MV504 are formed in the intermetallic dielectric layer 192. In more detail, the metal line ML31 contacts the metal vias MV31 and MV32. The metal line ML32 contacts the metal vias MV33 and MV34. The metal vias MV501 and MV502 contact the metal vias MV401 and MV402, respectively. The metal vias MV503 and MV504 contact the metal vias MV403 and MV404, respectively.

Refer to FIG. 28. An intermetallic dielectric layer 193 is formed over the intermetallic dielectric layer 192. Thereafter, metal vias MV601, MV602, MV603, and MV604 are formed in the intermetallic dielectric layer 193. In more detail, the metal vias MV601 and MV602 contact the metal vias MV501 and MV502, respectively. The metal vias MV603 and MV604 contact the metal vias MV503 and MV504, respectively. In some embodiments, the metal vias MV201, MV301, MV401, MV501, and MV601 together form the metal via MV41 discussed in FIGS. 2A to 2E. Similarly, metal vias MV202, MV302, MV402, MV502, and MV602 together form metal via MV42 discussed in FIGS. 2A to 2E.

Refer to FIG. 29. An intermetallic dielectric layer 194 is formed above the intermetallic dielectric layer 193. Then, a metal line ML41 and metal vias MV703 and MV704 are formed in the intermetallic dielectric layer 194. In more detail, the metal line ML41 contacts the metal vias MV601 and MV602. The metal vias MV703 and MV704 contact the metal vias MV603 and MV604, respectively.

Refer to FIG. 30. An intermetallic dielectric layer 195 is formed over the intermetallic dielectric layer 194. Then, metal vias MV803 and MV804 are formed in the intermetallic dielectric layer 195. In more detail, the metal vias MV803 and MV804 contact the metal vias MV703 and MV704, respectively. In some embodiments, the metal vias MV303, MV403, MV503, MV603, and MV703 together form the metal via MV51 discussed in FIGS. 2A to 2E. Similarly, the metal vias MV304, MV404, MV504, MV604, and MV704 together form the metal via MV52 discussed in FIGS. 2A to 2E.

Refer to FIG. 31. An intermetallic dielectric layer 196 is formed above the intermetallic dielectric layer 195. Then, metal lines ML51 and ML52 are formed in the intermetallic dielectric layer 196. In more detail, the metal lines ML51 and ML52 contact the metal vias MV803 and MV804, respectively. Thus, the SRAM element 10 is formed.

FIG. 32 is a three-dimensional diagram of an SRAM element of some embodiments of the present disclosure. FIG. 32 has substantially the same structure as FIG. 2C, similar elements have the same symbols, and will not be described again for the sake of brevity. The difference between FIG. 32 and FIG. 2C lies in the position of metal vias MV13 and MV14, wherein in FIG. 32, metal via MV13 contacts the bottom surface of source/drain contact 176, and metal via MV14 contacts the bottom surface of source/drain contact 174. In some embodiments, metal via MV11 is at least partially aligned with metal via MV13 along the X direction. Metal via MV12 is at least partially aligned with metal via MV14 along the X direction.

In some embodiments where transistors PU-1 and PU-2 are p-type transistors and transistors PD-1, PD-2, PG-1, and PG-2 are n-type transistors, metal lines ML11 and ML14 can be used together as the power line Vss of the SRAM element 10 (see FIG. 1 ). In other words, metal lines ML11 and ML14 can be regarded as the first and second parts of the power line Vss. Metal line ML12 can be used as the bit line BL of the SRAM element 10 (see FIG. 1 ), and metal line ML13 can be used as the bit line BLB of the SRAM element 10 (see FIG. 1 ). Alternatively, metal line ML12 can be used as the bit line BLB of the SRAM element 10, and metal line ML13 can be used as the bit line BL of the SRAM element 10.

In some embodiments where transistors PU-1 and PU-2 are n-type transistors and transistors PD-1, PD-2, PG-1, and PG-2 are p-type transistors, metal lines ML11 and ML14 can be used together as the power line Vdd of the SRAM element 10. In other words, metal lines ML11 and ML14 can be regarded as the first and second parts of the power line Vdd. Metal line ML12 can be used as the bit line BL of the SRAM element 10, and metal line ML13 can be used as the bit line BLB of the SRAM element 10. Alternatively, metal line ML12 can be used as the bit line BLB of the SRAM element 10. SRAM element 10, metal line ML13 can be used as the bit line BL of the SRAM element 10.

According to the above-mentioned embodiments, it can be seen that the present disclosure provides advantages in manufacturing integrated circuits. However, it should be understood that other embodiments may provide additional advantages, and not all advantages must be disclosed in this article, and all embodiments do not require specific advantages. The disclosed embodiments provide an SRAM structure with a CFET configuration, wherein transistor PU-1 and transistor PU-2 are stacked on transistor PD-1, transistor PG-1, transistor PD-2, transistor PG-2. In addition, signal lines (such as bit lines) and power lines (such as Vdd or Vss) are arranged below transistor PD-1, transistor PG-1, transistor PD-2, transistor PG-2 and signal lines (such as word lines). The power line (such as Vss or Vdd) is arranged above transistor PU-1 and transistor PU-2. This configuration can realize a small SRAM cell area by stacking transistors. The disclosed embodiments provide an SRAM structure with a staggered metal routing configuration to separate signal lines (e.g., bit lines BL, BLB) by power lines (e.g., Vdd or Vss). This configuration can prevent signal coupling between bit lines BL and bit lines BLB. Therefore, the voltage drop at bit line BLB can be reduced, which will increase the read speed and thus improve device performance.

According to some embodiments of the present disclosure, a memory element includes a substrate. A first pull-down transistor, a first gate transistor, a second pull-down transistor, and a second gate transistor are located at a first level above the substrate. A first pull-up transistor and a second pull-up transistor are located at a second level above the first level. A first power line, a first bit line, and a second bit line, the first power line including a first portion and a second portion separated from each other, wherein in a cross-sectional view, the second portion of the first power line is laterally located between the first bit line and the second bit line along a direction. A first conductive via contacts an upper surface of the first portion of the first power line and is electrically connected to a source/drain region of the first pull-down transistor. A second conductive via contacts an upper surface of the first bit line and is electrically connected to a source/drain region of the first gate transistor. A third via contacts the upper surface of the second portion of the first power line and is electrically connected to the source/drain region of the second pull-down transistor. A fourth via contacts the upper surface of the second bit line and is electrically connected to the source/drain region of the second gate transistor, wherein in a top view, the first via is laterally aligned with the fourth via along the direction, and the second via is laterally aligned with the third via along the direction.

In some embodiments, the memory device further includes a first semiconductor strip protruding from above the substrate and laterally disposed between a first portion of the first power line and a first bit line. A second semiconductor strip protruding from above the substrate and laterally disposed between a second portion of the first power line and a second bit line.

In some embodiments, the distance between the first bit line and the second portion of the first power line is shorter than the distance between the first portion of the first power line and the first bit line.

In some embodiments, the memory element further includes a liner lining the sidewalls and bottom surface of the first bit line.

In some embodiments, the memory device further includes a dielectric material lining the sidewalls of the first via, wherein the liner extends to the sidewalls of the dielectric material.

In some embodiments, the memory element further includes a word line electrically connected to the gate structure of the first gate transistor and the gate structure of the second gate transistor. A second power line electrically connected to the source/drain region of the first pull-up transistor and the source/drain region of the second pull-up transistor.

In some embodiments, the first pull-up transistor is vertically overlapped with the first pull-down transistor, and the second pull-up transistor is vertically overlapped with the second pull-down transistor.

According to some embodiments of the present disclosure, a memory device includes a substrate. A first pull-down transistor, a first gate transistor, a second pull-down transistor, and a second gate transistor are located at a first level above the substrate. A first pull-up transistor and a second pull-up transistor are located at a second level above the first level, wherein a gate structure of the first pull-up transistor vertically overlaps a gate structure of the first pull-down transistor, and a gate structure of the second pull-up transistor vertically overlaps a gate structure of the second pull-down transistor. A first power line, a first bit line, and a second bit line, wherein the first power line includes a first portion and a second portion separated from each other. A first conductive via contacts an upper surface of a first portion of the first power line and is electrically connected to a source/drain region of the first pull-down transistor. The second conductive via contacts the upper surface of the second portion of the first power line and is electrically connected to the source/drain region of the second pull-down transistor. The third conductive via contacts the upper surface of the first bit line and is electrically connected to the source/drain region of the first gate transistor. The fourth conductive via contacts the upper surface of the second bit line and is electrically connected to the source/drain region of the second gate transistor. The first liner, the second liner, the third liner, and the fourth liner are respectively attached to the sidewall of the first portion of the first power line, the sidewall of the second portion of the first power line, the sidewall of the first bit line, and the sidewall of the second bit line. The first dielectric material, the second dielectric material, the third dielectric material and the fourth dielectric material are respectively lined with the sidewalls of the first via hole, the sidewalls of the second via hole, the sidewalls of the third via hole and the sidewalls of the fourth via hole, wherein the first liner, the second liner, the third liner and the fourth liner are composed of materials different from the first dielectric material, the second dielectric material, the third dielectric material and the fourth dielectric material.

In some embodiments, the first liner, the second liner, the third liner, and the fourth liner extend to the sidewalls of the first dielectric material, the second dielectric material, the third dielectric material, and the fourth dielectric material, respectively.

In some embodiments, the first dielectric material, the second dielectric material, the third dielectric material, and the fourth dielectric material are separated from the first portion of the first power line, the second portion of the first power line, the first bit line, and the second bit line, respectively.

In some embodiments, the first substrate, the second substrate, the third substrate, and the fourth substrate extend to the upper surfaces of the first portion of the first power line, the second portion of the first power line, the first bit line, and the second bit line, respectively.

In some embodiments, the memory device further includes a semiconductor strip located on the substrate, wherein the first substrate and the third substrate contact the semiconductor strip.

In some embodiments, the second substrate contacts the third substrate.

In some embodiments, the first power line, the first bit line, and the second bit line are below the first level.

In some embodiments, the memory element further includes a word line electrically connected to the gate structure of the first gate transistor and the gate structure of the second gate transistor. A second power line electrically connected to the source/drain region of the first pull-up transistor and the source/drain region of the second pull-up transistor.

According to some embodiments of the present disclosure, a method includes forming a first semiconductor strip and a second semiconductor strip protruding above a substrate. Forming a first isolation structure, a second isolation structure, and a third isolation structure above the substrate, wherein the first semiconductor strip is located between the first isolation structure and the second isolation structure, and the second semiconductor strip is located between the second isolation structure and the third isolation structure. Patterning the second isolation structure and the third isolation structure to form a first trench and a second trench in the second isolation structure and the third isolation structure, respectively. Forming a first metal line and a second metal line in the first trench and the second trench, respectively. After forming the first metal line and the second metal line, patterning the first isolation structure and the second isolation structure to form a third trench and a fourth trench in the first isolation structure and the second isolation structure, respectively. A third metal line and a fourth metal line are formed in the third trench and the fourth trench, respectively. A first pull-down transistor electrically connected to the third metal line, a first gate transistor electrically connected to the first metal line, a second pull-down transistor electrically connected to the fourth metal line, and a second gate transistor electrically connected to the second metal line are formed. A first pull-up transistor and a second pull-up transistor are formed, vertically stacked above the first pull-down transistor, the first gate transistor, the second pull-down transistor, and the second gate transistor.

In some embodiments, the method further includes forming a first liner and a second liner in the first trench and the second trench, respectively, before forming the first metal line and the second metal line. Before forming the third metal line and the fourth metal line, forming a third liner and a fourth liner in the third trench and the fourth trench, respectively. Etching back the first metal line, the second metal line, the third metal line, and the fourth metal line, so that the upper surfaces of the first metal line, the second metal line, the third metal line, and the fourth metal line are lowered to a position lower than the top of the first liner, the second liner, the third liner, and the fourth liner.

In some embodiments, the method further includes forming a first via hole, a second via hole, a third via hole, and a fourth via hole after etching back the first metal line, the second metal line, the third metal line, and the fourth metal line, respectively contacting the first metal line, the second metal line, the third metal line, and the fourth metal line.

In some embodiments, the method further includes forming a first dielectric material, a second dielectric material, a third dielectric material, and a fourth dielectric material on the first metal line, the second metal line, the third metal line, and the fourth metal line, respectively, before forming the first via hole, the second via hole, the third via hole, and the fourth via hole, wherein the first via hole, the second via hole, the third via hole, and the fourth via hole are formed in the first dielectric material, the second dielectric material, the third dielectric material, and the fourth dielectric material, respectively.

In some embodiments, the first metal line serves as a first bit line, the second metal line serves as a second bit line, and the third metal line and the fourth metal line serve together as a first power line, wherein the method further includes forming a word line electrically connected to the first gate transistor and the second gate transistor. Forming a second power line electrically connected to the first pull-up transistor and the second pull-up transistor.

The foregoing summarizes the features of several embodiments so that those skilled in the art can better understand the state of the disclosure. Those skilled in the art should understand that they can easily use the disclosure as a basis for designing or modifying other processing procedures and structures to achieve the same purpose and/or achieve the same advantages of the embodiments described herein. Those skilled in the art should also recognize that such equivalent structures do not depart from the spirit and scope of the disclosure, and that various changes, substitutions and modifications can be made without departing from the spirit and scope of the disclosure.

10:SRAM components

152,154,156,158: Gate structure

171,172,173,174,175,176: Source/drain contacts

252,254: Gate structure

271,272,273,274: Source/drain contacts

D1,D2,D3:Distance

LV1: First level

LV2: Second level

ML11,ML12,ML13,ML14,ML31,ML32,ML41,ML51,ML52:Metal wire

MV11,MV12,MV13,MV14,MV21,MV22,MV23,MV31,MV32,MV33,MV34,MV41,MV42,MV51,MV52: Metal vias

PD-1, PD-2: Pull-down transistor/transistor

PG-1, PG-2: Gate transistor/transistor

PU-1,PU-2:Pull-up transistor/transistor

SD11,SD12,SD13,SD16,SD22,SD24: Source/Drain epitaxial structure

A-A,B-B: line

Claims (10)

A memory element comprises: a substrate; a first pull-down transistor, a first gate transistor, a second pull-down transistor and a second gate transistor, located at a first level above the substrate; a first pull-up transistor and a second pull-up transistor, located at a second level above the first level; a first power line, a first bit line and a second bit line, the first power line comprising a first portion and a second portion separated from each other, wherein in a cross-sectional view, the second portion of the first power line is laterally located between the first bit line and the second bit line along a direction; a first conductive hole, contacting a first portion of the first power line; a second conductive via contacting an upper surface of the first bit line and electrically connected to a source/drain region of the first gate transistor; a third conductive via contacting an upper surface of the second portion of the first power line and electrically connected to a source/drain region of the second pull-down transistor; and a fourth conductive via contacting an upper surface of the second bit line and electrically connected to a source/drain region of the second gate transistor, wherein in a top view, the first conductive via is laterally aligned with the fourth conductive via along the direction, and the second conductive via is laterally aligned with the third conductive via along the direction. The memory device as described in claim 1 further comprises: a first semiconductor strip protruding from above the substrate and laterally disposed between the first portion of the first power line and the first bit line; and a second semiconductor strip protruding from above the substrate and laterally disposed between the second portion of the first power line and the second bit line. A memory element as described in claim 1, wherein a distance between the first bit line and the second portion of the first power line is shorter than a distance between the first portion of the first power line and the first bit line. The memory device as described in claim 1 further comprises: a word line electrically connected to a gate structure of the first gate transistor and a gate structure of the second gate transistor; and a second power line electrically connected to a source/drain region of the first pull-up transistor and a source/drain region of the second pull-up transistor. A memory element comprises: a substrate; a first pull-down transistor, a first gate transistor, a second pull-down transistor and a second gate transistor, located at a first level above the substrate; a first pull-up transistor and a second pull-up transistor, located at a second level above the first level, wherein a gate structure of the first pull-up transistor vertically overlaps a gate structure of the first pull-down transistor, and a gate structure of the second pull-up transistor vertically overlaps a gate structure of the first pull-down transistor. a gate structure of the second pull-down transistor; a first power line, a first bit line, and a second bit line, wherein the first power line includes a first portion and a second portion separated from each other; a first conductive via contacting an upper surface of the first portion of the first power line and electrically connected to a source/drain region of the first pull-down transistor; a second conductive via contacting an upper surface of the second portion of the first power line and electrically connected to a source/drain region of the second pull-down transistor; a third conductive via contacting an upper surface of the first bit line and electrically connected to a source/drain region of the first gate transistor; a fourth conductive via contacting an upper surface of the second bit line and electrically connected to a source/drain region of the second gate transistor; a first liner, a second liner, a third liner and a fourth liner, respectively lining the sidewalls of the first portion of the first power line, the sidewalls of the second portion of the first power line, the sidewalls of the first bit line and the sidewall of the second bit line; and a first dielectric material, a second dielectric material, a third dielectric material and a fourth dielectric material, respectively lining the sidewall of the first via hole, the sidewall of the second via hole, the sidewall of the third via hole and the sidewall of the fourth via hole, wherein the first liner, the second liner, the third liner and the fourth liner are composed of materials different from the first dielectric material, the second dielectric material, the third dielectric material and the fourth dielectric material. A memory element as described in claim 5, wherein the first liner, the second liner, the third liner and the fourth liner extend to the sidewalls of the first dielectric material, the second dielectric material, the third dielectric material and the fourth dielectric material, respectively. A memory element as described in claim 5, wherein the first power line, the first bit line and the second bit line are lower than the first level. A method for forming a memory element comprises: forming a first semiconductor strip and a second semiconductor strip protruding above a substrate; forming a first isolation structure, a second isolation structure, and a third isolation structure above the substrate, wherein the first semiconductor strip is located between the first isolation structure and the second isolation structure, and the second semiconductor strip is located between the second isolation structure and the third isolation structure; patterning the second isolation structure and the third isolation structure to form a first trench and a second trench in the second isolation structure and the third isolation structure, respectively; forming a first metal line and a second metal line in the first trench and the second trench, respectively; after forming the first metal line and the second metal line, The first isolation structure and the second isolation structure are configured to form a third trench and a fourth trench in the first isolation structure and the second isolation structure, respectively; a third metal line and a fourth metal line are formed in the third trench and the fourth trench, respectively; a first pull-down transistor electrically connected to the third metal line, a first gate transistor electrically connected to the first metal line, a second pull-down transistor electrically connected to the fourth metal line, and a second gate transistor electrically connected to the second metal line are formed; and a first pull-up transistor and a second pull-up transistor are formed, which are vertically stacked above the first pull-down transistor, the first gate transistor, the second pull-down transistor, and the second gate transistor. The method as described in claim 8 further comprises: before forming the first metal line and the second metal line, forming a first liner and a second liner in the first trench and the second trench respectively; before forming the third metal line and the fourth metal line, forming a third liner and a fourth liner in the third trench and the fourth trench respectively; and etching back the first metal line, the second metal line, the third metal line and the fourth metal line so that the upper surfaces of the first metal line, the second metal line, the third metal line and the fourth metal line are lowered to a position lower than the top ends of the first liner, the second liner, the third liner and the fourth liner. The method as described in claim 9 further includes forming a first conductive hole, a second conductive hole, a third conductive hole and a fourth conductive hole after etching back the first metal line, the second metal line, the third metal line and the fourth metal line, respectively contacting the first metal line, the second metal line, the third metal line and the fourth metal line.

Images